106412 - DATA ANALYSIS

Subject type: Mandatory

Coordinator: Adso Fernández Baena

Trimester: First term

Credits: 6

The course introduces the student to the world of data analytics, with application to the analysis of video game data. Data analysis becomes a fundamental aspect of game development, in many ways:

• It helps to understand the user's behavior and to be able to adapt to it to improve the user's experience.
• You can identify types of users, by their behavior, by the type of strategies they use, or by the type of monetization they choose or the money they spend.
• Knowing how the player plays, if there are significant difficulties at certain points in the game or is too simple, the time they spend finishing a certain level or the playing time in each session, etc., are important data to be able to adjust the game in testing and balancing phases.
• You can try alternative versions of a given game and analyze which one is "most successful", according to the parameters you want to measure as successful (number of players, playing time, revenue it generates ...)
• Data analysis is also important for adjusting the monetization of a video game.

The subject is contextualized in the area of ​​Production and Business of the Degree in Design and Production of Video Games. The contents are based on a review of the most common metrics in video game design and monetization and make an introduction to inferential statistics and data analysis with machine learning methods. The R language is used throughout the course for the exercises and practical examples. The methodology combines master classes with exercises and practical activities. The evaluation activities are practical exercises and an analytical project that counts 60% of the mark and the remaining 40% corresponds to a final exam. 

The subject has no prerequisites.

In general, this subject contributes to the following learning outcomes specified for the subject to which it belongs (Production and Business):

• E15.2. Design the corresponding analyzes for a correct monitoring of the product once it is launched on the market.
• E15.3. Plan and develop the game data analysis process.
• E15.4. Interpret the results of the game analysis and design strategies to improve the game.

More specifically, at the end of the course, the student will be able to:

• Understand the need to analyze video game data.
• Explain video game analytics case studies.
• Apply video game analytics to different aspects of video game development, such as monetization, design, user capture, improving the gaming experience, and so on.
• Define relevant metrics, measure and analyze them.
• Explain what a hypothesis test is.
• Apply an appropriate hypothesis test.
• Explain the difference between descriptive statistics and inferential statistics.
• Explain the main approaches to data mining: regression, classification, grouping and association.
• Apply data mining algorithms for the extraction of useful information.
• Use R data analysis software.
• Make graphs that visually capture the information present in the data.
• Write data analytics reports at the executive and technical level.

The course is divided into theoretical sessions and practical sessions, corresponding to 4 hours and 2 hours per week respectively.

The work methodology combines:

• Master sessions, where the necessary concepts are explained.
• Readings of articles by the student, as complements of the contents seen in class.
• Video capsules, as a complement to the master classes.
• Carrying out practical exercises.

The content of the subject consists of the sections listed below:

1. Introduction to data analytics
   1. Importance of data analysis in video games
   2. What is data analysis?
   3. What is game analytics?
   4. Exercises and examples
2. Video game analysis metrics
   1. Types of metrics
   2. Game-specific metrics
   3. Population metrics
   4. Monetization metrics
   5. Marketing metrics
3. Introduction to the R tool
   1. Development environment R
   2. Data management in R
   3. Main orders
   4. Information display
4. Introduction to statistics
   1. Descriptive statistics
   2. Basic descriptive parameters
   3. Graphics
   4. Application of descriptive statistics to the analysis of video game metrics
5. Inferential statistics
   1. Introduction to hypothesis testing
   2. Hypothesis tests of a sample
   3. Two-sample hypothesis tests
   4. Application: A / B test of a video game design
6. Machine Learning
   1. What is machine learning?
   2. Main phases of a data mining process based on machine learning.
   3. Main approaches to machine learning: regression, classification, grouping.
   4. Application to video games
7. Reporting
   1. How to present data analytics information
   2. Drawing conclusions
8. Other visual data analysis tools: Tableau, Microsoft PowerBI.

The contents will be alternated with practical cases of application in order to see the usefulness of the contents that are treated throughout the subject.

The student will have to realize different activities along the asignatura:

• A1. Exercises - Analytical cases
• A2. Laboratory Practices - Video Game Data Analysis Project
• A4. Exam

The following details its focus and objectives.

A1. Exercises - Analytical cases

The aim of the practical exercises is that the student acquires the knowledge of the theoretical concepts seen in class and that he has agility in the use of the analytical tools that will be treated. These exercises aim to consolidate the work of the following skills:

• G1. Acquire advanced knowledge of data analytics through practical exercises.
• G2. Solve complex analytical problems.
• G3. Gather data and interpret it.
• E15. Design and implement quality assurance strategies, testing and data analysis of video games and interactive products.

These exercises are evidence for the achievement of the learning outcome E15.3 (plan and develop the game data analysis process).

A2. Laboratory practices: Video game data analysis project

The aim of the laboratory practices is that the student develops one or more cases of data analysis, where he will have to apply in an integral and grounded way the knowledge seen in class. The student will have to solve a simulation of a real case, limited in its complexity and volume of data, to make easier its management on the part of the student.

The following competencies will be developed in these projects:

• G5. Develop the learning skills necessary to acquire autonomy (because the projects are not supervised, but it is the student himself who must find the solutions to the projects, applying the concepts acquired).
• T1. Communicate in a third language, as at least one of the works must be written entirely in English.
• E15. Design and implement quality assurance strategies, testing and data analysis of video games and interactive products.
• G1. Acquire advanced knowledge of data analytics through practical exercises.
• G2. Solve complex analytical problems.
• G3. Gather data and interpret it.
• G4. Communicate to a specialized and non-specialized audience (as part of the project report, the executive report, must be written for non-specialized audiences and the other part of the report, the technical report, must to be written for specialized audiences).

and learning outcomes:

• E15.2. Design the corresponding analyzes for a correct monitoring of the product once it is launched on the market.
• E15.3. Plan and develop the game data analysis process.
• E15.4. Interpret the results of game analysis and design strategies to improve the game.

The student will carry out this team work (of two people, ideally) and will have to deliver a detailed report. This report shall contain an executive summary, a detailed report at the executive level, a technical report and an appendix with the data resulting from the analytical processes applied. The student will have the index of the work to deliver, as well as a rubric with the parameters of evaluation of the work.

A4. Final Exam

At the end of the course, each student will have to present to a final examination where will evaluate him of the contents seen along the asignatura. The exam is individual.

In this exam, the specific competences (E15), as well as the competences G5, G1, G2, G3 and part of G4, and the learning outcomes E15.2, E15.3 and E15.4 mentioned above will be assessed.

The evaluation of the subject is:

1. Practical exercises at home or in class: 30%
2. Laboratory practices (analytical project): 30%
3. Final exam: 40%

Continuous assessment activities must be delivered on time within the course specified. Beyond the specified deadlines, the student will not be able to deliver the activities of continuous evaluation, running the risk of suspending the subject for this reason. In the call for recovery it will not be possible to deliver the continuous assessment activities.

The following aspects must be carefully considered:

• Class attendance is mandatory, with a required minimum of 70% attendance.
• The minimum grade for the final exam is 4. If the student gets a lower grade, he will not average with the activities and will have to go to a resit exam. In case of going to recovery, the average will be calculated in the same way, substituting the mark of the exam by the mark of the recovery exam.
• The practical exercises must be delivered on time. Otherwise, they will count a 0 on the note.
• In the analytical practices (analytical project) there will be two delivery dates: the ordinary call and the extraordinary call (for exceptional cases). The analytical practices delivered in extraordinary call will have a maximum of 5. A maximum delivery date will be specified for the extraordinary call beyond which it will not be possible to deliver the internships and therefore will count as a 0. It is recommended that the student does not plan to deliver in the extraordinary call because it involves a decrease in the note.

Garcia-Ruiz, MA (2016). Games User Research. A Case Study Approach. CRC Press.

Wallner, G. (2019). Data Analytics Applications in Gaming and Entertainment. CRC Press.

Ugarte, MD, Militino, Ana F., & Arnholt, AT (2020). Probability and Statistics with R (2nd edition). CRC Press.

de Vries, A., & Meys, J. (2015). R for Dummies. John Wiley & Sons.

Brett Lanz (2013). Machine Learning with R. Learn how to use R to apply powerful machine learning methods and gain an insight into real-world applications. PACKT Publishing.

Magy Seif El-Nasr & Anders Drachen (2013). Game Analytics: Maximizing the Value of Player Data. Springer.

Witten, IH, Frank, E., & Hall, MA (2011). Data Mining. Practical Machine Learning Tools and Techniques. Third Edition. Morgan Kaufmann.

Bari, A., Chaouchi, M. & Jung, T. (2014). Predictive Analytics for Dummies. John Wiley and Sons.

Zumel, N. & Mount, J. (2014). Practical Data Science with R. Shelter Island: Manning.

Arun Sukumar, Lucian Tipi & Jayne Revill (2016). Applied Business Analysis. Available at: bookboon.com.

Brink, David (2010). Essentials of Statistics: Exercises. Available at: bookboon.com.